RE: OPINIONS WANTED: regexps in CSS? (Re: Suggestion for Attribute Selectors)

> -----Original Message-----
> From: www-style-request@w3.org [mailto:www-style-request@w3.org]On
> Behalf Of Victoria Rosenfeld
> Sent: Tuesday, March 10, 1998 12:47 PM
> To: Stephanos Piperoglou
> Cc: Bert Bos; www-style@w3.org
> Subject: Re: OPINIONS WANTED: regexps in CSS? (Re: Suggestion for
> Attribute Selectors)

> If "advanced" means: "users fluent in Perl" ... how many people do you
> think will be in this audience? The use of these REs seems to go counter
> to what the web is all about, and it *ain't* about having the "advanced"
> users be the only ones to use it to the fullest.

Who are you kidding? Any advanced user will have an edge over "less
advanced" users when using the Web. There is an advantage had in knowing how
the systems work that can't be attained any other way. Certainly the
maturation of the Web  has increased the number of facilities available to
less advanced users--and indeed, *all* users--but each of those easy-to-use
facilities has some not-so-easy-to-understand guts associated with it.

> Javascript seems like the way to go, in general.

JavaScript is an encumbered technology. The trademark is owned by Sun, and
the language seems to be maintained by Netscape. This is a non-starter.

ECMAScript, OTHO, is a possibility. I don't think this is a good idea, but
it's within the realm of possibility.

> Slightly complicated, but
> very powerful *and* can be implemented at it's basic level (mouse overs,
> popup windows, et all) fairly easily. Perl is a bitch to learn & I shudder
> at the thougt of having to become "advanced" enough to use (what is
> potentially) a great functionality such as CSSes.

Perl *is* a bitch to learn. But why do you think we need to add a scripting
language to CSS at all?

REs aren't really a bitch to learn--the principles are pretty simple. But
they are a bitch for most humans to parse in their minds. That's an
important point. I think CSS should remain a human-writable format, and
knowing REs shouldn't be part of the cost of entry. However, the power
afforded by REs cannot be ignored, nor can the relative ubiquity of REs in
the CS world.

I think it would be great to have REs available for style sheet authoring,
but they must not be required for using CSS. I'd suggest that anything that
can be expressed in a RE in CSS should be able to be expressed in other
means, even if they are much more verbose.

Braden McDaniel

Received on Tuesday, 10 March 1998 16:52:52 UTC